Model comparison
Claude 3.5 Haiku vs Claude 3 Opus
Claude 3.5 Haiku and Claude 3 Opus score almost the same on the Noometry Index (29.2 vs 29.5), so choose on price, context window or the category you care about most.
Last verified . 35 shared benchmarks.
Summary
- They share 35 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 4 categories and Claude 3 Opus in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Claude 3 Opus leads 24.5 to 18.7.
- The biggest single-benchmark swing is LiveBench Language: 35.4% for Claude 3.5 Haiku and 50.4% for Claude 3 Opus.
Side by side
| Claude 3.5 Haiku | Claude 3 Opus | |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 29.2 | 29.5 |
| Released | 2024-10-22 | 2024-02-29 |
| Weights | Proprietary | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 49 | 46 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Claude 3.5 Haiku: 32.9 (#265), Claude 3 Opus: 32.9 (#267)
| Benchmark | Claude 3.5 Haiku | Claude 3 Opus |
|---|---|---|
| WeirdML | 30.7% | 19.2% |
| BigCodeBench Instruct | 46.1% | 45.5% |
| LiveBench Coding | 51.4% | 38.6% |
| LMArena Coding | 1286 | 1264 |
| BigCodeBench Complete | 59% | 57.4% |
| Aider Polyglot | 28% | — |
| SciCode | 27.4% | — |
| CadEval | 32% | — |
| HumanEval+ | — | 77.4% |
| MBPP+ | — | 73.3% |
Agentic & Tool Use Claude 3.5 Haiku leads
Claude 3.5 Haiku: 28.0 (#95), Claude 3 Opus: 24.6 (#116)
| Benchmark | Claude 3.5 Haiku | Claude 3 Opus |
|---|---|---|
| Cybench | — | 10% |
| BALROG | 19.3% | — |
| METR Time Horizons | — | 29.5% |
Reasoning Claude 3.5 Haiku leads
Claude 3.5 Haiku: 17.7 (#290), Claude 3 Opus: 14.6 (#324)
| Benchmark | Claude 3.5 Haiku | Claude 3 Opus |
|---|---|---|
| LiveBench Reasoning | 28.1% | 40.6% |
| LMArena Hard Prompts | 1251 | 1245 |
| DTBench | 56.7% | 61.6% |
| LiveBench Data Analysis | 48.5% | 57.9% |
| Epoch Capabilities Index | 127.15 | 126.91 |
| LiveBench | 43.5% | 49.2% |
| SimpleBench | — | 23.5% |
| CritPt | 0% | — |
| Chess Puzzles | — | 5% |
| EnigmaEval | — | 0.8% |
| LMCA | — | 17% |
| ForecastBench | — | 58.4 |
| WinoGrande | — | 88.5% |
Math Too close to call
Claude 3.5 Haiku: 14.7 (#300), Claude 3 Opus: 14.8 (#299)
| Benchmark | Claude 3.5 Haiku | Claude 3 Opus |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 4.7% |
| LiveBench Math | 35.5% | 43.6% |
| LMArena Math | 1244 | 1273 |
| MATH Level 5 | 46.4% | 37.5% |
| Omni-MATH | 22.4% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Claude 3 Opus leads
Claude 3.5 Haiku: 18.7 (#281), Claude 3 Opus: 24.5 (#267)
| Benchmark | Claude 3.5 Haiku | Claude 3 Opus |
|---|---|---|
| GPQA Diamond | 38.1% | 47.2% |
| Confabulations | 36.7% | 22.7% |
| LMArena Expert | 1208 | 1223 |
| MMLU | 74.3% | 84.6% |
| SimpleQA Verified | — | 12.6% |
| MMLU-Pro | 60.5% | — |
| GPQA (HELM) | 36.3% | — |
Multimodal Too close to call
Claude 3.5 Haiku: 26.8 (#117), Claude 3 Opus: 27.1 (#116)
| Benchmark | Claude 3.5 Haiku | Claude 3 Opus |
|---|---|---|
| LMArena Vision | 1092 | 1023 |
| GeoBench | 34% | — |
Multilingual Claude 3 Opus leads
Claude 3.5 Haiku: 40.0 (#218), Claude 3 Opus: 41.4 (#207)
| Benchmark | Claude 3.5 Haiku | Claude 3 Opus |
|---|---|---|
| LMArena Non-English | 1238 | 1258 |
| LMArena Chinese | 1229 | 1248 |
| LMArena French | 1264 | 1275 |
| LMArena German | 1237 | 1258 |
| LMArena Japanese | 1175 | 1204 |
| LMArena Korean | 1173 | 1187 |
| LMArena Russian | 1253 | 1280 |
| LMArena Spanish | 1261 | 1246 |
Instruction Following Claude 3 Opus leads
Claude 3.5 Haiku: 62.9 (#234), Claude 3 Opus: 64.1 (#228)
| Benchmark | Claude 3.5 Haiku | Claude 3 Opus |
|---|---|---|
| LiveBench Instruction Following | 61.9% | 63.9% |
| LMArena Instruction Following | 1241 | 1248 |
| IFEval | 79.2% | — |
Long Context Too close to call
Claude 3.5 Haiku: 38.3 (#200), Claude 3 Opus: 38.2 (#202)
| Benchmark | Claude 3.5 Haiku | Claude 3 Opus |
|---|---|---|
| LMArena Longer Query | 1261 | 1259 |
Writing & Preference Claude 3 Opus leads
Claude 3.5 Haiku: 42.7 (#234), Claude 3 Opus: 47.2 (#213)
| Benchmark | Claude 3.5 Haiku | Claude 3 Opus |
|---|---|---|
| LMArena Text | 1255 | 1262 |
| LMArena Creative Writing | 1233 | 1235 |
| LMArena Multi-Turn | 1265 | 1275 |
| LiveBench Language | 35.4% | 50.4% |
| Short-Story Creative Writing | 73.5% | — |
| EQ-Bench Creative Writing | 1146 | — |
| WildBench | 76% | — |
Frequently asked questions
Is Claude 3.5 Haiku better than Claude 3 Opus?
Claude 3.5 Haiku and Claude 3 Opus score almost the same on the Noometry Index (29.2 vs 29.5), so choose on price, context window or the category you care about most.
Is Claude 3.5 Haiku or Claude 3 Opus better for coding?
They score almost the same on coding (32.9 vs 32.9); test both on your own repository before choosing.
How many benchmarks do Claude 3.5 Haiku and Claude 3 Opus share?
35 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and Claude 3 Opus has 46.